Re: [GT] Inconsistencies in date related options for scripts

["João Costa" <joaocosta AT zonalivre.org>]

Hi Thomas, I agree with your assessment, having consistent arguments for the
scripts is definitely a plus.  Some thoughts on the issues you raised:

2. Difference in treatment of timeframes

Checking all available timeframes allows scripts to handle cases where a
timeframe not stored in the datasource can be derived from a smaller
timeframe.  That's a nice feature, so I'd go for that option.

3. Differences in algorithm that determines first and last period

if only graphic.pl is different i guess it's easier to just change
graphic.pl.  Is there any significant advantage is using the algorithm in
graphic.pl ?  If not i'd just change graphic.pl to save time.

4. Differences on the default number of periods considered
>Some indicators (graphics.pl) base the default number of periods
>analyzed on --nb-item ...

This can be misleading if #2 above is implemented. In that scenario the
right way to do it would be to determine it based on the timeframe ratio.
However, keep in mind the timefrate ratio can be a double value and is even
infinite for the Tick timeframe, so I'd make sure that wouldn't be an issue.


5. Finally, display_indicator.pl supports --last-record. This would also
make sense for display_signal.pl

Sure does.



Since you have many of these changes implemented please submit them, they
definitely improve GT.



Cheers,
João